Low prevalence of arrhythmias in clinically stable COVID-19 patients.

BACKGROUND: No studies investigated the prevalence of arrhythmias among clinically-stable patients affected by COVID-19 infection.
METHODS: We assessed prevalence, type, and burden of arrhythmias, by a single-day snapshot in seven non-intensive COVID Units at a third-level center.
RESULTS: We enrolled 132 inhospital patients (mean age 65±14y; 66% males) newly diagnosed with COVID-19 infection. Arrhythmic episodes were detected in 12 patients (9%). In detail, 8 had atrial fibrillation, and 4 self-limiting supraventricular tachyarrhythmias. There were no cases of ventricular arrhythmias or new-onset atrioventricular blocks. In addition, we report no patients with QTc interval >450 ms.
CONCLUSIONS: Our single-day snapshot survey suggests that the prevalence of arrhythmias among clinically stable COVID-19 patients is low. In particular, no life-threatening arrhythmic events occurred.

INTRODUCTION

Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) is responsible for the current global health emergency. Among cardiovascular manifestations associated with the disease, arrhythmias have been reported. , However, two relevant biases involve previous studies addressing this topic: the under‐reporting of patients’ past arrhythmic history and major comorbities in a known association with arrhythmias; and the critical conditions of the most severe cases, admitted to intensive care units (ICU), leading to a possible overestimation of the arrhythmic burden. So far, no data are provided about clinically stable patients admitted to COVID units.

METHODS

We aimed at assessing, by a single‐day snapshot, the prevalence of arrhythmias among patients admitted to seven COVID units at a third‐level hub center. The focus of this study was restricted to clinically stable patients only, defined as those with no need for ICU stay. After designating referral physicians at each COVID unit, specific questionnaire (content reported in Table 1) was administered to all admitted patients in a single day. Anamnestic and clinical data, from admission to the study day, were collected by locally working personnel. All inpatients had baseline electrocardiogram (ECG). The warning cutoff for QTc interval prolongation was set at 500 ms, since sufficient to avoid life‐threatening arrhythmias. Continuous telemonitoring was performed in patients on CPAP. In the remaining cases, ECG was repeated only in the presence of new symptoms or alarm signs at routine daily check of parameters. All patients were under standard treatment including oral hydroxychloroquine 200 mg bid and azithromycin 500 MG IV daily. In compliance with the Declaration of Helsinki, our study was approved by the local Institutional Review Board.

TABLE 1

Baseline characteristics of the population

		Total	A+	A−
Feature	Units	n = 132	n = 12	n = 116
Age, y	Mean ± SD	65 ± 14	73 ± 10	64 ± 14
Caucasian	n (%)	116 (88)	12 (100)	100 (86)
CAD	n (%)	9 (7)	1 (8)	7 (6)
COPD	n (%)	8 (6)	2 (17)	5 (4)
Hypertension	n (%)	60 (45)	7 (58)	52 (45)
Diabetes	n (%)	26 (20)	1 (8)	24 (21)
Obesity	n (%)	19 (14)	1 (8)	17 (15)
History of AF	n (%)	16 (12)	2 (17)	10 (9)
≥1 RF	n (%)	81 (61)	9 (75)	68 (59)
IHSL	Median (IQR)	11 (6‐17)	19 (16‐23)	11 (6‐16)
Swab+	n (%)	125 (95)	10 (83)	111 (96)
CT+	n (%)	64 (48)	6 (50)	57 (49)
Swab+ CT+	n (%)	57 (43)	4 (33)	52 (45)
Swab+ CT−	n (%)	68 (52)	6 (50)	59 (51)
Swab− CT+	n (%)	7 (5)	2 (17)	5 (4)
Swab− CT−	n (%)	0 (0)	0 (0)	0 (0)
O2‐therapy	n (%)	94 (71)	11 (92)	79 (68)
CPAP	n (%)	50 (38)	6 (50)	41 (35)
IMV	n (%)	0 (0)	0 (0)	0 (0)
HCQ	n (%)	132 (100)	12 (100)	116 (100)
AZT	n (%)	132 (100)	12 (100)	116 (100)
Oral AAD	n (%)	18 (14)	8 (67)	10 (9)
Amiodarone	n (%)	14 (11)	8 (67)	6 (5)
Flecainide	n (%)	2 (2)	0 (0)	2 (2)
Sotalol	n (%)	2 (2)	0 (0)	2 (82)

Baseline characteristics of the population are shown, including comparison between patients with (A+) and without arrhythmias (A−). Patients with permanent AF (n = 4) were excluded by comparison.

Abbreviations: AAD, antiarrhythmic drugs; AF, atrial fibrillation; AZT, azythromicin; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CPAP, continuous positive airways pressure; CT, computed tomography; HCQ, hydrocychloroquine; IHSL, in‐hospital stay length; IMV, invasive mechanical ventilation; IQR, interquartile range; RF, risk factor.

RESULTS

Patient characteristics

We enrolled 132 patients (mean age 65 ± 14 years; 66% males) from seven COVID units. SARS‐CoV‐2 diagnosis was proved by pharyngeal swab (n = 68), typical abnormalities at chest CT scan (n = 7), or both (n = 57). Overall, 50 patients (38%) were in CPAP, and the remaining in lower grade oxygen therapy. Complete patient characteristics are reported in Table 1. Arrhythmic history included atrial fibrillation (AF) in 16 patients (12%; paroxysmal, n = 12; permanent, n = 4) and paroxysmal supraventricular tachycardia (PSVT) in one patient. Furthermore, three patients (2%) were pacemaker (PM) carriers. There was no history of ventricular arrhythmias in the whole cohort.

Documented arrhythmias

By the study day, new arrhythmic episodes during hospitalization were detected in 12 patients (9%). In detail, eight of 12 had AF (paroxysmal, n = 4; permanent, n = 4), excluding the four cases with permanent arrhythmia. Of them, six (75%) had CT scan‐proven diagnosis of SARS‐CoV‐2, and five (63%) were on CPAP. Of the cases with history of AF, only two of 12 (17%) had new episodes during hospitalization. Paroxysmal AF was self‐limiting in one case and interrupted by electrical or pharmacological (IV amiodarone) cardioversion in two and one cases, respectively.

The remaining four of 12 patients had atrial tachycardia (n = 3) or PSVT (n = 1). In all cases, a single and self‐limiting arrhythmic episode was reported. Overall, there were no cases of ventricular arrhythmias or new‐onset atrioventricular blocks. We report no patients with QTc interval >450 ms, including cases with ECG repeated (n = 23).

Overall, there were no differences in occurrence of arrhythmias between CT+ versus CT− (six of 12 vs 57 of 116, P = 1.000), swab+ versus swab− (10 of 12 vs 111 of 116, P = .130), and swab+ CT− versus CT+ (six of 63 vs six of 65, P = 1.000). As shown in Table 1, patients with arrhythmias had older age and longer in‐hospital stay, as compared to those without.

DISCUSSION

This study aimed at critically assessing both incidence and types of arrhythmias among stable patients affected by the SARS‐CoV‐2 infection. Differently from previous studies, we reported a lower incidence of arrhythmias, and no differences between swab+ patients and those with CT scan‐proven pneumonia or requiring CPAP for a more severe illness. Also, by no means a pathophysiologic link between arrhythmias and SARS‐CoV‐2 infection can be proven. In particular, as the most commonly documented arrhythmia, AF is nonspecific in critically ill patients on mechanical ventilation. Furthermore, patients undergoing AF were older, and most of them had at least one preexisting risk factor, including hypertension (Table 1). Finally, a minor quote of patients with known history of paroxysmal AF had recurrences during the COVID‐19 respiratory infection. Even in the single published case of biopsy‐proven myocarditis associated with SARS‐CoV‐2 respiratory infection, a single self‐limited episode of atrial arrhythmia was reported. Importantly, ventricular arrhythmias were completely absent in our cohort, even considering that all patients were on hydroxychloroquine and azythromicin, both associated with risk of QT interval prolongation. Therefore, our data suggest that the association with COVID‐19 infection and increased arrhythmic risk is low among patients outside ICU.

CONCLUSION

Our single‐day snapshot survey suggests that the prevalence of arrhythmias among clinically stable COVID‐19 patients is low. In particular, no life‐threatening arrhythmic events occurred, including CT+ and swab+ patients. All the reported episodes were supraventricular arrhythmias, always self‐limiting except for isolated cases of AF. While waiting for confirmatory data by larger studies, our findings suggest a low arrhythmic risk in clinically stable patients.

CONFLICT OF INTEREST

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.